Production of 4-hydroxybutyl acrylate and its reaction kinetics over Amberlyst 15 catalyst.

Yang, Jung-Il ; Cho, Soon-Haeng ; Kim, Hak-Joo 等

INTRODUCTION

In the coating industry, 2-HEMA (2-hydroxyethyl methacrylate) has been widely used as a coating agent. However, 4-hydroxybutyl acrylate (HBA) is known to be an environmentally benign product and it has better properties such as lustre-resistance, chemical-resistance and scratch-resistance than 2-HEMA. Thus, its consumption is expected to grow rapidly.

HBA can be produced from the esterification reaction of acrylic acid or methyl acrylic acid with 1,4-butanediol. Esterification is an acid-catalyzed reaction and liquid acid catalysts are frequently used to increase the reaction rate. Although sulphuric acid, hydrochloric acid, hydrofluoric acid and p-TSA possess strong catalytic activities, many side reactions accompany their use. They also cause corrosion and other environmental problems during the disposal of wastes. Significant research efforts have been devoted to find environmentally benign solid acid catalysts that can replace liquid acid catalysts. There are many advantages of solid acid catalysts, which are reduction of corrosion, ease of product recovery, less pollutant in waste stream and the repeated use of the catalyst.

The possible candidates of solid acid catalysts for esterification are solid oxides and ion exchanged resins (Chen et al., 1999). Solid oxides can be acidic salts of heteropolyacids, sulphated zirconia, titania-silica and silica-alumina. Ion exchanged resins with the acidic property are Amberlyst, DOWEX, Nafion, etc. Among these solid acid catalysts, ion exchanged resins are reported to be very active in the liquid-phase esterification reaction due to their special porous matrix and surface acid site features (Harmer and Sun, 2001). The catalysis by the ion exchanged resin in alkylation, transalkylation, isomerization, oligomerization, acylation, esterification and nitration is well reviewed by Harmer and Sun (2001). Amyl acetate preparation by DOWEX was demonstrated by Lee et al. (2000). N-butyl propionate production over Amberlyst was carried out by Liu and Tan (2001) and they showed that the activity of the Amberlyst catalyst was higher than that of HZSM5. Yadav and Mehta (1994) also proved that Amberlyst 15 was more active than sulphated zirconia, heteropolyacid and concentrated sulphuric acid.

The possible kinetic models for expressing esterification by solid acid catalysts are the quasi-homogeneous model, the Langmuir-Hinshelwood model, and the Eley-Rideal model (Lee et al., 2000; Liu and Tan, 2001; Xu and Chuang, 1996; Bart et al., 1996; Al-Jarallah et al., 1988). The quasi-homogeneous model is similar to the power law model of the homogeneous reaction, where the reaction rate is proportional to the concentration of reactants. The Langmuir-Hinshelwood model is derived on the basis that the rate-controlling step is the surface reaction of adsorbed species on a catalyst surface. The Eley-Rideal model is derived from the assumption that the rate-controlling step is the surface reaction between the adsorbed species on the catalyst surface and the species in the gas phase. The Eley-Rideal model was used to calculate the reaction rate of the n-butyl propionate production reaction over the Amberlyst catalyst (Liu and Tan, 2001). The heat of the reaction calculated from the model was very small and it was confirmed by almost isothermal reaction temperatures. Acetic acid esterification using Amberlyst 15 was also studied by Xu and Chuang (1996), and the quasi-homogeneous model was used to express the kinetic equation. Their kinetic equation included the stirrer speed, the reaction temperature, the reactant concentrations and the catalyst loading.

Acrylic acid and acrylate monomers are highly reactive. The high reactivity of the acrylic acid and the acrylate comes from unsaturated terminal sites of their backbone structures (Sartomer, 1990). Therefore, inhibitors, such as phenothiazine, hydroquinone (HQ), and hydroquinone mono methyl ether (MEHQ), are commonly used to prevent polymerization (Levy, 2003).

In this work, esterification of acrylic acid with 1,4-butanediol to produce 4-hydroxybutyl acrylate in liquid phase was carried our over several ion exchange resins. The reaction variables such as reactant concentrations, the catalyst amount and reaction temperatures were changed to provide necessary data for the kinetic expression. The quasi-homogeneous model was chosen to develop the complete kinetic equation of the esterification reaction to produce HBA.

ESTERIFICATION MECHANISM

Two reactions take place when acrylic acid (C[H.sub.2] = CH-COOH, AA) and 1,4-butanediol (HO-[(C[H.sub.2]).sub.4.]-OH, BD) react with each other by acid catalyst; esterification of AA with BD to form 4-hydroxybutyl acrylate (C[H.sub.2] = CH-COO-[(C[H.sub.2]).sub.4]-OH, HBA) and water (main reaction), and esterification of AA with the produced HBA to form 1,4-butanediol diacrylate (C[H.sub.2] = CH-COO-(C[H.sub.2])4 -OOC-HC=C[H.sub.2], BDA) and water (side reaction). Not only chemical structures of AA, BD, HBA and BDA, but also reaction mechanisms of the two reactions are described in Figures 1 and 2. As shown in Figures 1 and 2, the main reaction and side reaction are explained as follows.

Main reaction

* Protonation at the carbonyl oxygen of AA by acid catalyst: carbocation production;

* Attack at positive carbon by alcohol (BD): nucleophilic attack by hydroxyl group;

* Protonation at a hydroxyl then loss of water: stabilization of the reaction intermediate;

* HBA production.

Side reaction

* Protonation at the carbonyl oxygen of AA by acid catalyst: carbocation production;

* Attack at positive carbon by alcohol (HBA): nucleophilic attack by hydroxyl group;

* Protonation at a hydroxyl then loss of water: stabilization of the reaction intermediate;

* BDA production.

KINETIC MODEL

We used the liquid concentration instead of the activity of each component during the kinetic model development. In general, the activity is frequently used to account for the non-ideal effect of each component. For both the reaction of acetic acid and methanol to form methyl acetate and the hydrolysis reaction of methyl acetate over Amberlyst 15, the reaction rates calculated using activities rather than mole factions represented the experimental values better (Popken et al., 2000). The activity coefficients in the liquid phase can be calculated by the Non-Random-Two-Liquid (NRTL) model (Lee et al., 2000; Liu and Tan, 2001). In the absence of available data to calculate the activity coefficients using the NRTL, methods such as UNIversal QUAsi-Chemical (UNIQUAC) or UNIQUAC Functional-group Activity Coefficients (UNIFAC) are used to estimate activity coefficients (Bart et al., 1996; Popken et al., 2000; Rihko and Krause, 1995; Rehfinger and Hoffmann, 1990; Linnekoski et al., 1997). However, since the phase equilibrium data of our reactants and products are not available, we cannot use either NRTL model or UNIFAC method to estimate activities. Thus, we used liquid concentrations in our kinetic model development. In several cases, kinetic models were developed using the concentration instead of the activity (Yadav and Mehta, 1994; Xu and Chuang, 1996; Al-Jarallah et al., 1988; Lilja et al., 2002).

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

Two reactions take place when acrylic acid (AA) and 1,4-butanediol (BD) react with each other; esterification of AA with BD to form 4-hydroxybutyl acrylate (HBA) and water (Equation (1)), and esterification of AA with the produced HBA to form 1,4-butanediol diacrylate (BDA) and water (Equation (2)). HBA is the desired product.

AA + BD = HBA + [H.sub.2]O (1)

AA + HBA = BDA + [H.sub.2]O (2)

The HBA forming reaction can be divided into three individual steps described below.

AA + S = AA x S; - [r.sub.1] = [k.sub.1] [C.sub.AA] [C.sub.S] - [k.sub.-1] [C.sub.AAS] (3)

AA x S + BD = HBA x S + [H.sub.2]O; (4)

[-r.sub.2] = [k.sub.2] [C.sub.AAS][C.sub.BD] - [k.sub.-2] [C.sub.HBAS][C.sub.H2O] (4)

HBA x S = HBA + S; [-r.sub.3] = [k.sub.3][C.sub.HBAS] - [k.sub.-3] [C.sub.HBA][C.sub.S] (5)

If we assume that the reaction between the adsorbed AA and BD in the liquid phase (Equation (4)) is the rate-controlling step, the overall reaction rate becomes:

[-r.sub.a] = [-r.sub.2] = [k.sub.2][K.sub.1][C.sub.t]([C.sub.AA] [C.sub.BD]-[C.sub.HBA][C.sub.H2O]/[K.sub.eq,a])/ (1 + [K.sub.1][C.sub.AA] + [K.sub.-3][C.sub.HBA]) (6)

In Equation (6), [K.sub.eq,a] is the equilibrium constant of the esterification reaction of AA with BD (Equation (1)).

If we also assume the adsorption of AA and HBA on the catalyst surface is weak (very small [K.sub.1] and [K.sub.-3]), the denominator in the right-hand side of Equation (6) becomes 1 and the reaction rate can be expressed as:

[-r.sub.a] = [k.sub.a]([C.sub.AA][C.sub.BD] - [C.sub.HBA][C.sub.H2O]/ [K.sub.eq,a]), [k.sub.a] = [k.sub.2][K.sub.1][C.sub.t] (7)

The second BDA production reaction can also be divided into three individual steps described below.

AA + S = AA x S; [-r.sub.4] = [k.sub.4][C.sub.AA][C.sub.S] - [k.sub.-4] [C.sub.AAS] (8)

AA x S + HBA = BDA x S + [H.sub.2]O;

[-r.sub.5] = [k.sub.5][C.sub.AAS][C.sub.HBA] - [k.sub.-5][C.sub.BDAS] [C.sub.H2O] (9)

BDA x S = BDA + S; [-r.sub.6] = [k.sub.6][C.sub.BDA] - [k.sub.-6] [C.sub.BDAS][C.sub.S] (10)

Similarly to the HBA production reaction, the rate of overall reaction of BDA production can be expressed as the following with the assumption that the reaction between the adsorbed AA and HBA in a liquid phase (Equation (9)) is rate controlling.

[-r.sub.b] = [-r.sub.5] = [k.sub.5][K.sub.4][C.sub.t]([C.sub.HBA] [C.sub.AA] - [C.sub.BDA][C.sub.H2O]/[K.sub.eq,b])/ (1 + [K.sub.4][C.sub.AA] + [K.sub.-6][C.sub.BDA]) (11)

In the above Equation (11), [K.sub.eq,b] is the equilibrium constant of the esterification reaction of AA with HBA (Equation (2)). If the adsorption of AA and BDA on the catalyst surface is assumed to be weak ([K.sub.4] and [K.sub.-6] are very small), the denominator in the right-hand side of Equation (11) becomes 1 and the reaction rate is reduced to be:

[-r.sub.b] = [k.sub.b]([C.sub.HBA][C.sub.AA] - [C.sub.BDA][C.sub.H2O]/ [K.sub.eq,b]), [k.sub.b] = [k.sub.5][K.sub.4][C.sub.t] (12)

As shown in Equations (7) and (12), both rate equations are similar to the reaction rates derived from the quasi-homogeneous model, which can also be deduced from the Eley-Rideal model when the adsorptions of the reacting species are weak.

The rates of formation or disappearance of each species can be expressed with either [-r.sub.a], [-r.sub.b] or both.

[-r/sub.AA] = [-dC.sub.AA]/dt = ([-r.sub.a])+([-r.sub.b]) = [k.sub.a]([C.sub.AA][C.sub.BD] - [C.sub.HBA][C.sub.H2O]/[K.sub.eq,a]) + [k.sub.b]([C.sub.HBA][C.sub.AA] - [C.sub.BDA][C.sub.H2O]/[K.sub.eq,b]) (13)

[-r.sub.BD] = [-dC.sub.BD]/dt = ([-r.sub.a]) = [k.sub.a]([C.sub.AA] [C.sub.BD] - [C.sub.HBA][C.sub.H2O]/[K.sub.eq,a]) (14)

[r.sub.HBA] = [dC.sub.HBA]/dt = ([-r.sub.a])-([-r.sub.b]) = [k.sub.a]([C.sub.AA][C.sub.BD] - [C.sub.HBA][C.sub.H2O]/[K.sub.eq,a]) [-k.sub.b]([C.sub.HBA][C.sub.AA]- [C.sub.BDA][C.sub.H2O]/[K.sub.eq,b]) (15)

[r.sub.BDA] = [dC.sub.BDA]/dt = ([-r.sub.b]) = [k.sub.b]([C.sub.HBA] [C.sub.AA] - [C.sub.BDA][C.sub.H2O]/[K.sub.eq,b]) (16)

Integration of Equations (14) and (16) will result in integration values, [I.sub.a] and [I.sub.b], and they are:

[integral][-dC.sub.BD]/[C.sub.AA][C.sub.BD] - [C.sub.HBA][C.sub.H2O] / [K.sub.eq,a] = [I.sub.a] = [k.sub.a]t (17)

[intergral][dC.sub.BDA]/[C.sub.HBA][C.sub.AA] - [C.sub.BDA][C.sub.H2O] / [K.sub.eq,b] = [I.sub.b] = [k.sub.b]t (18)

In above equations, ka and kb are functions of the reaction temperature and the catalyst concentration. They can be expressed as follows:

[k.sub.a] = [k.sub.a,0] W exp(-[DELTA][E.sub.a]/RT) (19)

[k.sub.b] = [k.sub.b,0] W exp(-[DELTA][E.sub.b]/RT) (20)

EXPERIMENTAL

Chemicals

Acrylic acid (99%), 1,4-butanediol (99+%), 4-hydroxybutyl acrylate (96%) and 1,4-butanediol diacrylate (90%) were obtained from Aldrich Chemical Co. Phenothiazine (98% +) as the inhibitor was also supplied from Aldrich Chemical Co.

Catalysts

Commercial ion exchanged resins were tested as catalysts. Amberlyst 15 and Amberlyst 35 were supplied by Rohm and Hass Co. and DOWEX HCR-S(E) was obtained from DOW Co.

Experimental Procedure

Esterification was carried out using a three-neck flask (150 ml) with each port serving as a condenser, a thermometer well and a sampling port. Figure 3 shows schematic diagram of the reactor. The amount of total reactants was kept constant at 11.65 ml with varying ratios of AA to BD. Unless specified, the inhibitor amount was fixed to 17.5 mg (0.15 wt.%) (Yokoyama, 2000). A summary of the experiments is given in Table 1. After charging the reactants and the inhibitor into the flask, the flask was heated to the reaction temperature. When the desired temperature was attained, the reaction was initiated by immersing the catalyst into liquid reactants. The reaction temperature was changed from 100[degrees]C to 120[degrees]C . The reactants were mixed by a magnetic stirrer, and the stirring speed was changed to check the external mass transfer limitation. The sample amount taken by a syringe to analyze the liquid composition was 0.1 [micro]l. The concentrations of reactants and products were analyzed by using a gas chromatograph (DsChrom6200, Donam Co., South Korea) with [N.sub.2] as a carrier gas. The GC was equipped with a flame ionization detector and a column (AT-capillary column, Alltech Co.). The oven temperature was set at 250[degrees]C, which is higher than the boiling points of both reactants and products. The main product of the reaction was 4-hydroxybutyl acrylate ([CH.sub.2] = CH-COO-([C[H.sub.2]).sub.4]-OH, HBA), and the by-product was 1,4-butanediol diacrylate (C[H.sub.2] = CH-COO-[(C[H.sub.2]).sub.4]-OOC-HC = C[H.sub.2], BDA). The yield and selectivity of HBA are defined as follows:

HBA yield (%) = (moles of HBA produced)/(moles of initial BD) x 100 (21)

HBA selectivity (%) = (moles of HBA produced)/(sum of moles of HBA and (22) BDA) x 100

RESULTS AND DISCUSSION

Catalyst Screening

As a preliminary experiment, three representative solid acids were tested as catalysts for HBA production and the catalysts were [Cs.sub.2.5][H.sub.0.5][PW.sub.12][O.sub.40] ([Cs.sub.2.5]), Amberlyst 15 and Nafion-H. [Cs.sub.2.5] was selected due to its super acidity among solid oxides. Amberlyst 15 and Nafion-H were selected among organic resins because they took very different acid amount and surface area. Table 2 shows surface area and acid amount of solid acids, and Figure 4 represents HBA yield and selectivity over the three representative solid acids. As shown in Figure 4, Amberlyst 15 showed higher activities than [Cs.sub.2.5] and Nafion-H, which was strongly ascribed to its higher acid amount (4.9 mmol/g) compared to the others.

[FIGURE 3 OMITTED]

Three different ion exchanged resins with the similar acid amount were tested as the catalysts for esterification of AA with BD; Amberlyst 15, Amberlyst 35 and DOWEX HCR-S(E). The reaction was carried out at 110[degrees]C. The yield and the selectivity of HBA for each catalyst are shown in Figure 5. Amberlyst 15 and Amberlyst 35 showed higher activities than DOWEX HCR-S(E). As listed in Table 3, DOWEX HCR-S(E) has almost the same acid site concentration as Amberlyst 15. Their pore structures are different from each other. Both Amberlyst 15 and Amberlyst 35 are porous materials with the average pore diameters of 250 [Angstrom] and 300 [Angstrom], respectively, and furthermore the pores are continuously open (Rohm and Hass, 1996). However, DOWEX HCR-S(E) is a gel-type cation resin and its pore structure is microporous (DOW, 1996). It must be easier for reactants (AA and BD) to diffuse into the pores of Amberlyst 15 and Amberlyst 35 than than those of DOWEX HCR-S(E). Thus, the low catalyst activity of DOWEX HCR-S(E) seems to be due to its smaller pore size. Xu and Chuang (1996) also reported that Amberlite IR-120 Plus showed a low activity for acetic acid esterification because of its microporous pore structure. They also reported that Amberlyst 15 and Amberlyst 35 showed high activities due to larger pores.

To be more specific, the activity of Amberlyst 15 was slightly higher than that of Amberlyst 35. Amberlyst 35 was also considered as a good catalyst for the esterification, but the moisture content of Amberlyst 35 was slightly higher than that of Amberlyst 15 as shown in Table 3. Therefore, it was considered that its inferior activity compared to Amberlyst 15 was ascribed to its higher moisture contents. Yadav and Mehta (1994) reported that Amberlyst 15 was highly active compared to other solid acids in producing phenethyl acetate and cyclohexyl acetate. Patwardhan and Sharma (1990) also reported that Amberlyst 15 was the most active in esterification of carboxylic acids by olefins. We also chose Amberlyst 15 as the most suitable catalyst for the HBA production and the rates of the esterification reaction between AA and BD over Amberlyst 15 were measured.

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

External Mass Transfer Effect

The stirrer speed of the batch reactor was varied from 300 rpm to 750 rpm to confirm the existence of the external mass transfer limitation of esterification of AA by BD over Amberlyst 15. As shown in Figure 6, the production of HBA was independent of the stirrer speed. Thus, the external mass transfer was not rate controlling at a stirrer speed higher than 300 rpm. Chakrabarti and Sharma (1993) reported that the overall rate of the reaction catalyzed by ion exchange resins was not affected by the external diffusion unless the viscosity of reactants was too high or the agitation speed was too low. We set the stirrer speed constant at 750 rpm for other runs of the rate measurement.

Internal Mass Transfer Effect

Yadav and Mehta (1994) calculated the effectiveness factor of the Filtrol-24 catalyst for producing phenethyl acetate and of the Amberlyst catalyst for cyclohexyl acetate production. They found the effectiveness factor was 1 for both reactions. There were no intraparticle diffusional resistance for their catalysts of the average particle size of 660 [micro]m. Xu and Chuang (1996) also reported that the average effectiveness factor of the commercial Amberlyst 15 was 0.9 for methyl acetate production. We used Amberlyst 15 of a commercial type and their average particle size was 750 [micro]m. We presumed that the internal mass transfer resistance did not exist in all of the runs in this work.

[FIGURE 6 OMITTED]

[FIGURE 7 OMITTED]

Equilibrium Constant

Equilibrium constants for HBA production ([K.sub.eq,a]) and BDA production ([K.sub.eq,b]) can be expressed by equilibrium concentrations of the reaction mixture at equilibrium.

[K.sub.eq.a] = [C.sub.HBA][C.sub.H2O]/([C.sub.AA][C.sub.BD]) (23)

[K.sub.eq.b] = [C.sub.BDA][C.sub.H2O[/([C.sub.AA][C.sub.HBA]) (24)

The equilibrium constants at reaction temperatures, 100[degrees]C, 110[degrees]C and 120[degrees]C were measured. The reaction equilibrium was reached in 1 to 3 days depending on the reaction temperature. Figure 7 shows that [K.sub.eq.a] and [K.sub.eq.b] decrease slightly with an increase in reaction temperature.

Effect of Reactants Mole Ratio

The mole ratio of AA to BD was varied from 1:1 to 2.33:1 to assess its effect on HBA yield, BDA yield, and HBA selectivity. The total amount of reactants was 151.24 mmol (11.65 ml) at 120[degrees]C with a catalyst loading of 200 mg. The activities are presented in Table 4. As described in Table 4, the reaction time to take the maximum yields of HBA was decreasing with the mole ratio, but the maximum amount of produced HBA was also decreasing. On the other hand, the reaction rate was so slow when the reactants mole ratio of AA to BD was 1:1. When the reactants mole ratio increased to 2.33:1, the reaction time for taking the maximum yield of HBA was decreased and the maximum produced amount of HBA was also decreased because the smaller amount of BD was used and reaction rate of BDA was accelerated by excessive amount of AA. Therefore, the reactants mole ratio of AA to BD was fixed to 1.85:1 considering the yield of HBA, the produced amount of HBA, and the reaction time.

Reaction Rate Constant

Integration values ([I.sub.a] and [I.sub.b]) were calculated at different reaction times using the reaction data to obtain reaction rate constants and an example is shown in Figure 8. As can be seen in Figure 8, the plots of [I.sub.a] and [I.sub.b] vs. time yielded straight lines. The results shown in Figure 8 indicate that the esterification reactions of HBA and BDA productions catalyzed by Amberlyst 15 follow quasi-homogeneous model. Thus, Equations (13), (14), (15) and (16) are adequate to describe the reaction kinetics. The reaction rate constants were obtained from slopes of plots shown in Figure 8. For example, at the reaction temperature of 110[degrees]C and the catalyst weight of 200 mg, the rate constants are as follows:

[k.sub.a] = 1.0924 x [10.sup.-3] [mol.sup.-1] ml [min.sup.-1] (25)

[k.sub.b] = 2.1559 x [10.sup.-4] [mol.sup.-1] ml [min.sup.-1] (26)

Effect of Temperature

The reaction temperature was varied from 100[degrees]C to 120[degrees]C and reaction rates were measured. From the initial slopes of the lines shown in Figure 9, it can be seen that the reaction rate increased with the temperature. At 120[degrees]C, however, the concentration of HBA reached a maximum after 3 h of the reaction because the produced HBA was consumed by BDA production at a high rate.

Activation energies were obtained by plotting logarithm of rate constants vs. 1/T. As shown in Figure 10, the plots are straight lines, indicating that the internal diffusion resistance is not significant. From the slopes of the lines, the activation energies for the reactions (Equations (1) and (2)) are calculated to be 58.3 kJ/mol and 86.7 kJ/mol, respectively. The large values of the activation energies support our assumption that the reaction of adsorbed AA on the catalyst and BD in the liquid phase is the rate-controlling step.

[FIGURE 8 OMITTED]

[FIGURE 9 OMITTED]

Effect of Catalyst Concentration

The catalyst-loading amount was varied from 50 mg to 200 mg and the reaction was carried out at a temperature of 110[degrees]C. Figure 11 shows that the reaction rate constants increased linearly with the catalyst loading. The fact that the rate constant was close to zero at the zero catalyst loading indicates that the resin catalyst played a crucial role in esterification of AA by BD. Upon increasing the catalyst loading, the total number of active sites increases linearly, results in an increase of the reaction rate constants.

Effect of Inhibitor

To determine the effect of inhibitor, the inhibitor amount was also changed from 10 mg to 35 mg. Figure 12 shows HBA yield and selectivity depending on the inhibitor amount. As shown in Figure 12, the reaction activities were not influenced by the amount of inhibitor. However, polymerization occurred when there was no inhibitor addition in the esterification. Therefore, it was confirmed that although the inhibitor amount from 10 mg to 35 mg (0.08 wt.% ~ 0.28 wt.%) was not critical to the reaction, its addition was indispensable.

Kinetic Equation

Based on the results obtained above, the reaction rate of HBA production in the esterification reaction using Amberlyst 15 is expressed as the following equation:

[dC.sub.HBA]/dt = [k.sub.a]([C.sub.AA][C.sub.BD] - [C.sub.HBA][C.sub.H2O]/[K.sub.eq.a]) - [k.sub.b]([C.sub.HBA][C.sub.AA] - [C.sub.BDA][C.sub.H2O]/[K.sub.eq,b]) (15)

where the rate constants are as follows:

[k.sub.a] = [k.sub.a,0] W exp (-[DELTA][E.sub.a]/RT), [k.sub.a,0] = exp(15.49) [ml.sup.2]/(mol [g.sub.cat] min), [DELTA][E.sub.a] = 58.3 kJ/mol (27)

[k.sub.b] = [k.sub.b,0] W exp (-[DELTA][E.sub.b]/RT), [k.sub.b,0] = exp(22.80) [ml.sup.2]/(mol [g.sub.cat] min), [DELTA][E.sub.b] = 86.7 kJ/mol (28)

[FIGURE 10 OMITTED]

[FIGURE 11 OMITTED]

[FIGURE 12 OMITTED]

The equilibrium constants at different temperatures are shown in Figure 7.

With the above complete kinetic equation, the change of components' concentrations with time can be calculated and one example is shown in Figure 13. As can be seen, the kinetic equations developed in this work predicted the experimental results precisely.

Catalyst Reusability

Three repeated runs were carried out with 200 mg of the Amberlyst 15 catalyst. Each run took 9 h at 110[degrees]C. As shown in Figure 14, the rate of HBA formation of each run was almost same, indicating that the deactivation of the Amberlyst 15 catalyst did not take place during the three repeated runs. The difference between the results with the fresh catalyst and those of the reused catalyst was very small. Thus, it is assured that the catalyst activity was maintained well enough to obtain the meaningful kinetic data.

[FIGURE 13 OMITTED]

[FIGURE 14 OMITTED]

CONCLUSIONS

Esterification of acrylic acid with 1,4-butanediol produces 4-hydroxybutyl acrylate (HBA), an environmentally benign coating agent. The by-product, 1,4-butanediol diacrylate (BDA), is also formed by the reaction between acrylic acid and HBA. The Amberlyst 15 catalyst was superior to other ion exchanged resin catalysts such as Amberlyst 35 and DOWEX HCR-S(E). The external mass transfer resistance did not exist at the stirring speed higher than 300 rpm. The internal mass transfer resistance was insignificant for the catalyst of particle size of 750 [micro]m. The rates of esterification reactions of HBA and BDA formations were well expressed by the quasi-homogeneous model. The reaction rate constants were obtained by varying the reaction conditions such as the concentration, the temperature, and the catalyst-loading amount. The activation energies were 58.3 kJ/mol and 86.7 kJ/mol for HBA production and BDA production, respectively. The complete kinetic equation was developed based on the quasi-homogeneous model and it predicted experimental results well.

ACKNOWLEDGMENT

The financial support from the Korea Ministry of Commerce, Industry and Energy (A2C-07-02) is gratefully acknowledged.

NOMENCLATURE

[C.sub.i] concentration of
 component i in mixture
 (mol [ml.sup.-1])

[C.sub.iS] surface concentration of sites
 occupied by component i (mol
 [ml.sup.-1])

[C.sub.S] surface concentration of vacant
 sites (mol [ml.sup.-1])

[C.sub.t] total concentration of surface
 sites (mol [ml.sup.-1])

[E.sub.a], [E.sub.b] activation energy (kJ [mol.sup.-1])

[I.sub.a], [I.sub.b] integration values defined in
 Equations (17) and (18)
[k.sub.1], [k.sub.2], [k.sub.-2], rate constants (mol-1 ml
[k.sub.-3], [k.sub.4], [k.sub.5], [min.sup.-1])
[k.sub.-5], [k.sub.-6]

[k.sub.-1], [k.sub.3], [k.sub.-4], rate constants ([min.sup.-1])
[k.sub.6]

[k.sub.a0], [k.sub.b0] pre-exponential constants
 ([mol.sup.-1] [ml.sup.2]
 [min.sup.-1] [g.sub.cat.sup.-1])

[k.sub.a], [k.sub.b] rate constants ([mol.sup.-1]
 ml [min.sup.-1])

[K.sub.eq,a], [K.sub.eq,b] reaction equilibrium constants (-)

[K.sub.i], [K.sub.-i] surface reaction equilibrium
 constants ([mol.sup.-1] ml)

[-r.sub.i] rate of reaction for component i
 (mol [ml.sup.-1] [min.sup.-1])

R gas constant, 8.314 J [mol.sup.-1]
 [K.sup.-1]

S surface active site

t time (min)

T temperature (K)

W catalyst loading ([g.sub.cat] [ml.sup.-1])

Abbreviations

AA acrylic acid

BD 1,4-butanediol

HBA 4-hydroxybutyl acrylate

BDA 1,4-butanediol diacrylate

Manuscript received February 3, 2006; revised manuscript received July 11, 2006; accepted for publication September 13, 2006.

REFERENCES

Al-Jarallah, A. M., M. A. B. Siddiqui and A. K. K. Lee, "Kinetics of Methyl Tertiary Butyl Ether Synthesis Catalyzed by Ion Exchange Resin," Can. J. Chem. Eng. 66, 802-807 (1988).

Bart, H. I., W. Kaltenbrunner and H. Landschutzer, "Kinetics of Esterification of Acetic Acid with Propyl Alcohol by Heterogeneous Catalysis," Int. J. Chem. Kinet. 28, 649-656 (1996).

Chakrabarti, A. and M. M. Sharma, "Cationic Ion Exchange Resins as Catalyst," React. Polym. 20, 1-45 (1993).

Chen, X., Z. Xu and T. Okuhara, "Liquid Phase Esterification of Acrylic Acid with 1-Butanol Catalyzed by Solid Acid Catalysts," Appl. Catal. A: Gen. 180, 261-269 (1999).

DOW Company, "Dowex Monosphere Ion Exchange Resins Data Sheet," Midland, MI, U.S.A. (1996).

Harmer, M. A. and Q. Sun, "Solid Acid Catalysis using Ion-Exchane Resins," Appl. Catal. A: Gen. 221, 45-63 (2001).

Lee, M. J., H. T. Wu and H. M. Lin, "Kinetics of Catalytic Esterification of Acetic Acid and Amyl Alcohol over Dowex," Ind. Eng. Chem. Res. 39, 4094-4099 (2000).

Levy, L. B., "Inhibition of Acrylic Acid Polymerization by Phenothiazine and P-Methoxyphenol. II. Catalytic Inhibition by Phenothiazine," J. Polym. Sci. 30(4), 569-576 (2003).

Lilja, J., J. Aumo, D. Y. Murzin, P. Maki-Arvela, M. Sundell, K. Ekman, R. Peltonen and H. Vainio, "Kinetics of Esterification of Propionic Acid with Methanol over a Fibrous Polymer-Supported Sulphonic Acid Catalyst," Appl. Cat. A: Gen. 228, 253-267 (2002).

Linnekoski, J. A., A. O. Krause and L. K. Rihko, "Kinetics of the Heterogeneously Catalyzed Formation of tert-Amyl Ethyl Ether," Ind. Eng. Chem. Res. 36, 310-316 (1997).

Liu, W. T. and C. S. Tan, "Liquid-Phase Esterification of Propionic Acid with n-Butanol," Ind. Eng. Chem. Res. 40, 3281-3286 (2001).

Patwardhan, A. A. and M. M. Sharma, "Esterification of Carboxylic Acids with Olefins using Cation Exchange Resins as Catalysts," React. Polym. 13, 161-176 (1990).

Popken, T., L. Gotze and J. Gmehling, "Reaction Kinetics and Chemical Equilibrium of Homogeneously and Heterogeneously Catalyzed Acetic Acid Esterification with Methanol and Methyl Acetate Hydrolysis," Ind. Eng. Chem. Res. 39, 2601-2611 (2000).

Rehfinger, A. and U. Hoffmann, "Kinetics of Methyl Tertiary Butyl Ether Liquid Phase Synthesis Catalyzed by Ion Exchange Resin--I. Intrinsic Rate Expression in Liquid Phase Activities," Chem. Eng. Sci. 45, 1605-1617 (1990).

Rihko, L. K. and A. O. I. Krause, "Kinetics of Heterogeneously Catalyzed tert-Amyl Methyl Ether Reactions in the Liquid Phase," Ind. Eng. Chem. Res. 34, 1172-1180 (1995).

Rohm and Haas Company, "Ion Exchange Resin Data Sheet," Philadelphia, PA, U.S.A. (1996).

Sartomer Company, "The Effects Polymerization Inhibitors Have on Acrylate Monomers and Formulations," Pennsylvania, PA, U.S.A. (1990).

Yadav, G. D. and P. H. Mehta, "Heterogeneous Catalysis in Esterification Reactions: Preparation of Phenethyl Acetate and Cyclohexyl Acetate by Using a Variety of Solid Acidic Catalysts," Ing. Eng. Chem. Res. 33, 2198-2208 (1994).

Yokoyama, "Process for Producing Hydroxyalkyl Monoacrylate Using Stannoxane Catalysts," U.S.A. Patent 6,087,528 (2000).

Xu, Z. P. and K. T. Chuang, "Kinetics of Acetic Acid Esterification over Ion Exchange Catalysts," Can. J. Chem. Eng. 74, 493-500 (1996).

Jung-Il Yang (1), Soon-Haeng Cho (1), Hak-Joo Kim (1), Hyunku Joo (1), Heon Jung (1) and Kwan-Young Lee (2) *

(1.) Korea Institute of Energy Research, Daejeon 305-343, South Korea

(2.) Department of Chemical and Biological Engineering, Korea University, Seoul 136-701, South Korea

* Author to whom correspondence may be addressed. E-mail address: kylee@korea.ac.kr

Table 1. Summary of the experiments

 Reactants
 Stirrer mole ratio
Run speed
number [rpm] AA BD

Set 1 1 300 1.85 1
 2 450 1.85 1
 3 750 1.85 1

Set 2 4 750 1 1
 5 750 1.85 1
 6 750 2.33 1

Set 3 7 750 1.85 1
 8 750 1.85 1
 9 750 1.85 1

Set 4 10 750 1.85 1
 11 750 1.85 1
 12 750 1.85 1
 13 750 1.85 1

Set 5 14 750 1.85 1
 15 750 1.85 1
 16 750 1.85 1
 17 750 1.85 1

 Reaction Catalyst
 Temp. weight Inhibitor
Run
number [[degrees][mg] [mg]

Set 1 1 110 200 18
 2 110 200 18
 3 110 200 18

Set 2 4 120 200 18
 5 120 200 18
 6 120 200 18

Set 3 7 100 200 18
 8 110 200 18
 9 120 200 18

Set 4 10 120 50 18
 11 120 100 18
 12 120 150 18
 13 120 200 18

Set 5 14 100 200 0
 15 100 200 10
 16 100 200 18
 17 100 200 35

Table 2. Surface area and acid amount of solid
acid catalysts (Chen et al., 1999)

Catalysts Acid Surface
 amount area
 (mmol/g) [(m.sub.2]/g)
Solid oxides
[Cs.sub.2.5][H.sub.0.5]
 [PW.sub.12][O.sub.40] 0.15 116
S[O.sub.4.sub.2-]/Zr[O.sub.2] 0.20 93
H-ZSM-5 0.39 403
Si[O.sub.2]-[Al.sub.2][O.sub.3] 0.35 546

Organic resins
Amberlyst 15 4.90 50
Nafi on-H 0.80 0.02
DOWEX HCR-S(E) 4.80 0.60 (a)

(a) The value was determined by Nitrogen adsorption
(ASAP2010, Micrometrics, U.S.A.)

Table 3. Typical physical and chemical properties
of catalysts (Rohm and Haas, 1996; DOW, 1996)

Properties Amberlyst 15 Amberlyst 35 DOWEX-HCR-S(E)

Physical form opaque opaque light yellow,
 spherical spherical translucent
 beads beads spherical beads

Ionic form [H.sup.+] [H.sup.+] [H.sup.+]

Concentration of 4.9 5.4 4.8
acid site [meq/g]

Moisture content,
[mass%] 53 56 52~56

Particle size [mm] 0.35~1.20 0.4~1.25 0.3~1.2

Porosity [ml/g] 0.3 0.35 0.000077 (a)

Average pore 25 30 0.45 (a)
diameter [nm]

Surface area 45 44 0.60 (a)
[[m.sup.2]/g]

Maximum operating
temperature
[[degrees]C] 120 140 120

(a) The value was determined by Nitrogen adsorption
(ASAP2010, Micrometrics, U.S.A.)

Table 4. Effect of reactants mole ratio on
the esterification of AA with BD

 Mole
 ratio Max. HBA Max. HBA Reaction
 of AA yield produced BDA yield HBA sel. time (a)
 to BD [mol%] [mmol] [mol%] [mol%] [min]

1 : 1 52.2 39.5 13.9 78.9 356

1.85 : 1 52.7 28 21.1 71.5 215

2.33 : 1 52.4 23.8 14.5 78.3 110

(a) The value means the reaction time in which a maximum
yield of HBA can be obtained in the batch reactor

The reaction conditions; total amount of reactants
(151.24 mmol), 120[degrees]C